Fetal intestinal lactase develops relatively late in gestation and studies of breath hydrogen excretion have suggested that greater than 66% of dietary lactose may reach the colon in some premature infants. Based on these studies one might presume that clinical intolerance and inefficient energy absorption might be associated with the ingestion of the relatively large amounts of lactose present in human milk or in formulas containing lactose as the sole carbohydrate. Thus, special premature formulas have been developed which contain decreased lactose with the carbohydrate replaced in part by glucose polymer. However, our recent studies of premature infants suggested that less than 10% of lactose and/or glucose polymer energy was actually excreted in premature infants who were 28-32 wk. gestation at birth and 12-30 days of age when studied. If our studies as well as the breath hydrogen studies are valid and representative of what occurs in most premature infants, then one might speculate that colonic bacterial flora serve a ruminant-type function by fermenting lactose and producing short-chain fatty acids which can be absorbed in the colon. The specific aims of this proposal are: to determine fecal excretion of carbohydrate-derived energy and breath hydrogen excretion in premature infants fed lactose or lactose-plus-glucose polymer and to determine the fecal excretion of carbohydrates and short-chain fatty acids. These studies may provide information relevant to the development of appropriate formulas for such infants and also improve our understanding of the nutritional functions of the colon and the colonic bacterial flora in premature infants.